What is [ChatGPT](https://openai.com/blog/chatgpt/)? Before jumping to our topic, we can look at how ChatGPT introduces itself:

\
"ChatGPT is a large language model that was trained by OpenAI. It is designed to be able to generate human-like text in response to prompts given to it, and it is capable of doing so in a variety of different styles and formats. It is based on the GPT-3 (Generative Pretrained Transformer 3) architecture, which is a type of machine learning model that uses deep learning algorithms to generate text. ChatGPT is designed to be able to generate conversational text that is similar to what a human might say in response to a given prompt, making it well-suited for applications such as chatbots and natural language processing tasks."

\
What do you think? This sounds like a natural conversation. In fact, ChatGPT is a new AI-driven chatbot that can answer some questions and even write a paragraph of essays. When we search on Google for an answer, the search engine tries to retrieve webpages with higher matching scores to the questions. We need to check each webpage for a satisfactory answer. As humans write the content of webpages, the answer can be a good candidate. Unlike Google search, ChatGPT is trained with all webpage contents crawled from the internet. It learns the style of answering a question. Meanwhile, we can consider the content of an answer as the summarization of the major answers online. Hence, sometimes the answer can be wrong even though it reads like someone confidently answers a question. It is still too early to say it can answer questions with strong logic.

\
​As we are a computer hardware company, let us start to “interview” ChatGPT with questions on GPU graphics cards. From the answers, we believe that in the future, we can have a smart chatbot online to do customer service since most answers sound reasonable.

\
**Question: Is it a good idea to buy a used GPU?**

**ChatGPT:** It can be a good idea to buy a used GPU, but it's important to be cautious and do your research. Here are a few things to consider when buying a used GPU:


1. Condition: Make sure the GPU is in good working condition. Check for any visible damage, such as bent pins or cracks in the PCB.
2. Compatibility: Make sure the GPU is compatible with your computer's motherboard and power supply. Check the manufacturer's website for the minimum requirements for the GPU.
3. Performance: Check the specs of the used GPU and compare it to a new one. You may be able to find a used GPU that performs just as well as a new one, but at a lower price.
4. Warranty: Consider whether the used GPU comes with a warranty or return policy. Some sellers may offer a limited warranty or return policy, which can give you some peace of mind.

Overall, it's important to do your research and be cautious when buying a used GPU. It can be a good way to save money, but it's important to make sure you're getting a quality product.

\
**Q: What kinds of GPUs are good for 4K PC gaming, and what kinds of GPUs are good for AI computing?**

**ChatGPT:** For 4K PC gaming, you'll want a GPU with a high amount of memory and a fast clock speed. Some examples of GPUs that are good for 4K gaming include the NVIDIA GeForce RTX 3080, the AMD Radeon RX 6800 XT, and the NVIDIA GeForce RTX 3090. These GPUs offer a high level of performance and can handle the demands of 4K gaming.

For AI computing, you'll want a GPU with a high amount of memory and a fast clock speed, as well as support for parallel processing. Some examples of GPUs that are good for AI computing include the NVIDIA Tesla V100, the NVIDIA Quadro GV100, and the NVIDIA GeForce RTX 8000. These GPUs are specifically designed for AI and machine learning applications, and offer a high level of performance for these types of tasks.

It's worth noting that the specific GPU you choose will depend on your specific needs and budget. There are many other factors to consider, such as the amount of memory, the number of CUDA cores, and the memory bandwidth. It's important to do your research and choose the GPU that best meets your needs.

\
**Question: Please recommend some good GPUs for 4K PC gaming in the year 2022**

**ChatGPT:** In 2022, some good GPUs for 4K PC gaming are likely to include:


1. NVIDIA GeForce RTX 3080: This GPU offers excellent performance and is capable of running most games at 4K resolution with high graphics settings. It features 10GB of memory and 8704 CUDA cores.
2. AMD Radeon RX 6800 XT: This GPU also offers excellent performance and is capable of running most games at 4K resolution with high graphics settings. It features 16GB of memory and 4608 stream processors.
3. NVIDIA GeForce RTX 3090: This is the most powerful GPU currently available, and is ideal for 4K gaming. It features 24GB of memory and 10496 CUDA cores, making it capable of running even the most demanding games at 4K resolution with high graphics settings.

It's worth noting that the specific GPU you choose will depend on your budget and specific needs. There are many other factors to consider, such as the amount of memory, the number of CUDA cores, and the memory bandwidth. It's important to do your research and choose the GPU that best meets your needs.

\
**Question: How about the year of 2023?**

**ChatGPT:** It's difficult to predict exactly what the best GPUs for 4K PC gaming in 2023 will be, as new products are constantly being released and technology is constantly improving. However, some factors that are likely to influence the performance of GPUs in 2023 include the number of CUDA cores, the amount of memory, and the clock speed.

In general, you'll want to look for a GPU with a high number of CUDA cores, a large amount of memory, and a fast clock speed in order to get the best performance for 4K gaming. It's also a good idea to keep an eye on the latest GPU releases from companies like NVIDIA and AMD, as they often release new products that offer improved performance and features.

Ultimately, the best GPU for 4K gaming in 2023 will depend on your budget and specific needs. It's important to do your research and choose the GPU that best meets your needs.

\
**Questions: How many GPUs have been employed to train ChatGPT?**

**ChatGPT:** I'm sorry, but I don't have information about the specific hardware used to train ChatGPT. ChatGPT is a large language model developed by OpenAI, and it's likely that it was trained using a combination of CPU and GPU hardware. Training large language models like ChatGPT typically requires a lot of computational resources, including a large number of CPUs and GPUs. However, I don't have specific details about the hardware used to train ChatGPT.

\
**Some of the author’s thoughts:** ChatGPT will deeply affect many areas, such as all QA websites, HowTo websites, and even online search and programming. Some people have demonstrated that ChatGPT can generate some basic AI framework to finish a simple AI task. It will change education in the future too. Students are not trained to memory knowledge, but how to ask the right questions, which is a very important capability. ChatGPT can also help small businesses. It can be employed as an online chatbot to answer questions from customers. For example, as an ITAD company, BuySellRam.com receives customer queries daily, so we may leverage an AI chatbot to help customers to get a quote or to solve common computer problems.

Sell Computer CPU, Memory, GPU, SSD to BuySellRam.com

Read My Stories

Too Long; Didn't Read

Get hands-on learning from ML experts on Coursera 

Asking ChatGPT to Recommend a Graphic Card

Too Long; Didn't Read

@jimbsr

RELATED STORIES